MOBILE

Multi-Omics Binary Integration via Lasso Ensembles (MOBILE)

The MOBILE pipeline finds context-specific association networks. It integrates multi-omics datasets in a data-driven, biologically-structured manner. The gene-level association networks are used to nominate differentially enriched pathways. More info and application examples can be found here in the MOBILE paper.

Dependencies

• MATLAB
• glmnet package
• RStudio and R (for image quantification functions used in the paper)

Instructions

1. Clone this repository from the command-line using

   git clone --recursive https://github.com/cerdem12/MOBILE.git

2. Make sure all the folders are added to the MATLAB path.

3. Run the scripts in order:

   MOBILE_runATACseqRNAseq.m + MOBILE_runRNAseqRPPA.m => MOBILE_summarize.m => MOBILE_select.m

Model testing and performance

The MOBILE simulation of RPPA-RNAseq inference takes around 2-3 second per run (10000 instances are run in total: ~8 hours) including the save function in a normal desktop/laptop.

The RNAseq-ATACseq inference simulations were run on Clemson University Palmetto HPC and took around 8 hours per 1000 iteration of the 10000 instances (sources used per batch job: number of nodes=1, number of CPUs=40, memory=360gb). Ten batch jobs were run in parallel and the results were concatanated offline afterwards.

Tested environments include:

• Ubuntu 18.04, Intel Core i7 3930 CPU @ 3.20 GHz, 32 GB DDR3, Nvidia GTX 690 GPU
• Windows 10 Education, Intel Core i5-3470 CPU @ 3.20 GHz, 8.00 GB RAM, Nvidia GTX 650 GPU, 64-bit operating system
• Windows 10 Pro, Intel Core i7-8550U CPU @ 2.00 GHz, 16.00 GB RAM, Intel UHD 620 GPU, 64-bit operating system

Citation

Erdem C, Gross SM, Heiser LM*, Birtwistle MR* (2023). MOBILE pipeline enables identification of context-specific networks and regulatory mechanisms. Nature Communications. 14, 3991. doi: https://doi.org/10.1038/s41467-023-39729-2